• Certain individually stressful events have been observed to increase the psychological distress of persons affected; reduced psychological distress following other events has been attributed to collective processes, including increased group cohesion. These possibilities are investigated by contrasting reported symptom levels of 938 adults interviewed before, during, and after a racial riot. White suburbanites interviewed after the riot and urban black women interviewed during the riot report significantly fewer psychological symptoms.
Hypotheses of seasonal symptom changes, sampling biases, and the absence of symptom changes among relatively unimpaired respondents are rejected, suggesting that reductions in symptom level are associated with the riot. Serious methodological problems are raised by our finding that such events may substantially affect not only rates but also patterns of reported psychological symptoms obtained through epidemiological studies.